SOURCE-LINKED INTELLIGENCE

When Retrieval Helps: Selective Retrieval for Single-Turn Mental-Health QA

arXiv · AI, language, vision and robotics · article · Sep 3, 2026 · UTC

Retrieval-augmented generation (RAG) can improve the specificity and grounding of large language model responses, but its effect is not uniformly beneficial in single-turn mental-health question answering, where user queries often combine emotional distress, treatment concerns, and safety-sensitive needs. We study when retrieval helps or hurts mental-health QA, and whether a lightweight selective retrieval policy can better control this trade-off. We operationalize retrieval need using three draft-conditioned utility dimensions: psychoeducational need, coping need, and response specificity, to

Read original source ↗ Open in workspace

recordType

paper

region

Global